Mammographic density is well accepted as a risk factor for breast cancer. Many drawbacks exist for mammographic density measurements. Mammography is a projection 2D image, not a 3D image. Mammography suffers from imperfect calibration of parameters and variation in film exposure factors. Changes in exposure, compression, and processor chemistry may influence the background density of the film and potentially the measured breast density in each study, making comparison between subjects or longitudinally unreliable. Density measured on breast MRI has a great potential to provide a sensitive, quantitative method for evaluating breast density. MRI provides a great soft tissue contrast distinguishing between fibroglandular and fatty tissues;also it provides the 3-dimensional breast coverage;thus is an optimal imaging modality for evaluating fibroglandular density. Despite several studies used MRI aiming for developing automated computer-aided mammographic density analysis, there is little effort to develop a reliable method for monitoring the change breast density over time for individual woman. A quantitative MRI-based measure of "percent fibroglandular density" may be used to evaluate the changes induced by the chemoprevention drug. Secondary chemoprevention is commonly used for high-risk women who already had diagnosis of breast cancer. Incidence of breast cancer in the contralateral breast increases after treatment of breast caner. The cumulative incidence at 20 years was 15.4%. Tamoxifen was proven effective, and prescribed widely to pre-menopausal patients diagnosed with estrogen receptor positive breast cancer. Despite the promising results, whether to offer chemoprevention, or which drug to give, is solely based on population study results. There is not an individualized method that can be used for a woman to determine whether the chemoprevention is working on her, thus is beneficial over the possible side effects and risks. In this application we propose to use our developed 3D MRI-based quantitative breast density evaluation method to test the specific hypothesis that secondary chemoprevention with tamoxifen (for pre-menopausal women) will reduce the breast density compared to controls. We will also follow all enrolled subjects for future events of breast cancer development. When the data is available we will be able to directly test the hypothesis that reduced density is correlated with lower future cancer events. Two aims are proposed in this study 1) Aim-1: To evaluate the changes of breast density in pre-menopausal women receiving tamoxifen vs. controls using MRI-based analysis. 2) Aim-2: To obtain mammograms of all patients for density analysis, and compare the results to that based on MRI in terms of measurement variation and changes of density between tamoxifen vs. controls.